From: Tanaka Akira Date: 2007-12-31T16:49:17+09:00 Subject: Re: multibyte strings & bucket-of-bytes efficiency under 1.9.0 In article <14554987.post@talk.nabble.com>, Brent Roman writes: > Is it true that the slowness of string manipulations in Ruby 1.9 > is due to the fact that a String's encoding may be changed > at any time? (I really have not investigated this in detail) I think it is possible to optimize ASCII-8BIT string as Ruby 1.8 with small overhead. Ruby can use byte oriented fast implementation after testing the string is ASCII-8BIT. I don't think the test is slow. I guess the problem is a performance bug. For example, nobu fixed a performance bug of Array#join before Ruby 1.9.0-0. The bug caused that install-doc spends 40 minutes on my machine. I don't surprise if there are other performance bugs. So what we need is a (small) script which reproduce the bug. -- Tanaka Akira